Michael Barone thinks the Obama administration gave out stimulus money to state governments as a pay off to public employee unions:

Why did districts with state capitals rake in so much money? Because one-third of the stimulus dollars went to state and local governments. The obvious purpose was to shield public employee union members from the economic distress and uncertainty faced by so many Americans in the private sector—to pay off the unions (most of whose members are public employees) which contributed something like $400 million to Democrats in the 2008 cycle. I suppose this has some stimulative effect on the macroeconomy, though perhaps not as much as would have come to sending stimulus checks directly to Democratic donors. But it leaves Democrats vulnerable to the charge that they are favoring public sector employees (and public sector union leaders) over the far larger number of Americans who depend on the private sector.
In this recession we have lost about 7 million private sector jobs, while job loss in the public sector has been just about zero. I suppose [Nate] Silver might argue that pumping money into the public sector is a good thing because it maintains desperatelty [sic] needed public services and countercyclical spending like unemployment benefits. But does he really want to defend the proposition that you can't cut even one public sector job without damaging the public?

Sine it's pretty clear that Barone does not understand the theory he's arguing against, let me sum it up. Keynesian economics suggests that, in a severe recession, the government wants to pump demand into the economy through deficit spending. The federal government does this in part through "automatic stabilizers" -- during a recession, the decline in income and profits decreases tax revenue, and the increased need for food stamps, jobless benefits and other programs increases. States, however, are required to balance their budgets. Since revenues are falling and expenditures rising automatically, this forces them to cut spending and/or raise taxes automatically just to stay level. In other words, state fiscal policy has a large pro-cyclical effect, deepening the recession. The states lay off police officers, and then those officers stop going out to eat, forcing the local restaurant to lay off waitresses. Etc.

The purpose of providing aid to state and local governments is to help soften the pro-cyclical effect. Even with that aid, state government jobs have decreased by 5%, and local government jobs by 4%, over the last year.

The ideal fiscal policy during a massive economic crisis would be to temporarily increase the number of government employees, in order to soak up some of the enormous slack in the labor market. Mainstream economists believe aid to state and local governments is one of the most efficient ways to stimulate the economy. (But -- a ha! -- many economists work for universities, which are part of the public sector, so obviously they're trying to pay off the public employee unions, too.)

Barone notes derisively that the net effect of public sector jobs "has been about zero" -- right! That's because the increase in federal jobs has mostly compensated for the decrease in state and local jobs. Barone seems unable to understand this policy in macroeconomic terms, focusing on the question of fairness in apportioning job losses between the public and private sector. I'm not saying he's wrong because he disagrees with Keynesian economics. I'm saying he's wrong because he betrays no understanding of the rationale for the policy he's disparaging.